Biochemical characterization of the native α-carbonic anhydrase purified from the mantle of the Mediterranean mussel, Mytilus galloprovincialis

Abstract A α-carbonic anhydrase (CA, EC 4.2.1.1) has been purified and characterized biochemically from the mollusk Mytilus galloprovincialis. As in most mollusks, this α-CA is involved in the biomineralization processes leading to the precipitation of calcium carbonate in the mussel shell. The new enzyme had a molecular weight of 50 kDa, which is roughly two times higher than that of a monomeric α-class enzyme. Thus, Mytilus galloprovincialis α-CA is either a dimer, or similar to the Tridacna gigas CA described earlier, may have two different CA domains in its polypeptide chain. The Mytilus galloprovincialis α-CA sequence contained the three His residues acting as zinc ligands and the gate-keeper residues present in all α-CAs (Glu106-Thr199), but had a Lys in position 64 and not a His as proton shuttling residue, being thus similar to the human isoform hCA III. This probably explains the relatively low catalytic activity of Mytilus galloprovincialis α-CA, with the following kinetic parameters for the CO2 hydration reaction: kcat = 4.1 × 105 s−1 and kcat/Km of 3.6 × 107 M−1 × s−1. The enzyme activity was poorly inhibited by the sulfonamide acetazolamide, with a KI of 380 nM. This study is one of the few describing in detail the biochemical characterization of a molluskan CA and may be useful for understanding in detail the phylogeny of these enzymes, their role in biocalcification processes and their potential use in the biomimetic capture of the CO2.


Introduction
Mollusk shell is composed of about 95% calcium carbonate (CaCO 3 ) and 5% of organic components 1 . The shell is formed by three elements secreted by the mantle (the dorsal body of the animal): a) a thin outer periostracum; b) a middle prismatic, which is a crystalline form of calcium carbonate; c) an inner calcareous (nacreous) layer appearing with a vividness texture or iridescent mother-of-pearl (nacre), depending on the species [2][3][4] . Prismatic and nacreous layers are in the forms of calcium carbonate crystal; but, the two layers display a crystal polymorphism because the prismatic layer develops calcite and the nacreous layer forms aragonite 5 . The shell grows in circumference by the addition of material from the edge of the mantle, while grows in thickness by deposition from the general mantle surface. Calcium for shell growth is obtained from diet and/or from seawater, whereas carbonate is originated from the concentration of CO 2 /bicarbonate present in the tissue of the organism 6 . The interconversion of CO 2 and HCO 3 À is balanced naturally to maintain the equilibrium between dissolved inorganic carbon dioxide (CO 2 ), carbonic acid (H 2 CO 3 ), bicarbonate (HCO 3 À ) and carbonate (CO 3 2À ) [7][8][9] . The spontaneous hydration/dehydration reaction: CO 2 þ H 2 O … HCO 3 À þ H þ is very slow at physiological pH. In this context, a family of metalloenzymes, named carbonic anhydrase (CA; EC 4.2.1.1) assumes a pivotal role in the biomineralization process of the mollusks and is essential for the shell development 7,8,[10][11][12][13][14][15][16] .
CAs, in fact, have the primary function to strongly accelerate the interconversion of CO 2 and HCO 3 À playing an important role during the calcium carbonate formation in mollusks, such as acidbase regulation, calcification and mineralization 17,18 . It has been reported in literature that intracellular or cytoplasmic membrane bound CAs are essential for cytoplasmic acid/base balance and for transport mechanisms of CO 2 and carbonate ions; different CA isoforms has been identified in the calcifying epithelia of the mollusks 17,19 ; CA activity has been reported in the extracellular skeletal matrix of the cnidarian exoskeleton, crustacean calcium storage concretions, fish otolith and molluskan shell 17,19,20 ; moreover, nacrein, the protein involved in the nacreous layer formation of shell and pearl, identified for the first time in the Japanese pearl oyster Pinctada fucata, function has a CA because it has a CA domain in its amino acid sequence 21 .
The CA superfamily includes seven distinct classes known as the a, b, c, d, f, g and h. In addition to biomineralization, these enzymes are involved in many physiologic processes, such as photosynthesis, respiration, CO 2 transport, as well as metabolism of xenobiotics (e.g., cyanate in Escherichia coli) [13][14][15][22][23][24][25][26][27][28][29][30][31] . Some of the catalytically active aand h-CAs can also catalyze the hydrolysis of esters, such as 4-nitrophenyl acetate (4-NpA) 32 . However, no esterase activity was detected so far for enzymes belonging to the other five classes (b-, c-, d-, fand g-CAs) 33 . The a-, b-, d-, gand, perhaps h-CAs use Zn(II) ions at the active site, the c-CAs are probably Fe(II) enzymes (but they are active also with bound Zn(II) or Co(II) ions) [34][35][36][37][38][39][40][41] , whereas the f-class CAs are cambialistic enzymes, active both with Cd(II) or Zn(II) bounded within the active site in order to perform the physiologic reaction catalysis [42][43][44] . The metal ion from the enzyme active site is coordinated by three His residues in the a-, c-, dand h-classes, by one His and two Cys residues in band f-CAs or by two His and one Gln residues in g-class with the fourth ligand being a water molecule/ hydroxide ion acting as nucleophile in the catalytic cycle of the enzyme 8,12,13,[45][46][47] . In metazoans, CAs belong predominantly to the a-CA family, but recently b-CA family members have been identified 17 . In the last years, CAs have acquired a great importance in biotechnological applications, such as in the achievement of an artificial respiration system, selective biosensors for metal ions, and in the carbon capture process (CCP) 48,49 . In the CCP context, a number of CO 2 sequestration methods have been proposed in order to capture CO 2 using different types of CA enzymatic bioreactors 48,50 .
In the present study, we characterized and determined the kinetic constants of the CA purified from the mantle tissue of the bivalve Mediterranean mussel, Mytilus galloprovincialis. This CA has been purified by ammonium sulfate precipitation and ionexchange chromatography followed by affinity chromatography. From the determination of its N-amino terminal sequence the purified metalloenzyme it has been assigned to the a-class of the CA superfamily. The kinetic study of the active a-CA in the mantle of the adult specimens of Mytilus galloprovincialis may provide further information on the physiological role and function of this metalloenzymes in the process of biomineralization of the bivalves. This is one of the few contributions, apart the characterization of the a-CA from Tridacna gigas, in which molluskan CAs are investigated in detail at the biochemical level 51 .

Chemicals
All the chemicals were commercial products of the purest quality and purchased from Sigma. Immobilion-P membranes were from Perkin-Elmer (Waltham, MA).

Animals
Adult specimens of Mytilus galloprovincialis were collected in the proximity of Gulf of Naples. The mussels were maintained in seawater at a temperature of 4 C. Bivalve mantels were quickly removed and frozen at À70 C.

Enzyme purification
All the purification steps were carried out at a temperature of 4 C. Approximately, 20 g of mussel mantles were homogenized in 50 ml of 20 mM Tris-HCl buffer pH 8.3 containing 10 À3 M PMSF, 10 À3 M benzamidine and 2 Â 10 À3 M EDTA. The homogenate was centrifuged twice for 30 min at 12,000 Â g, and the resulting supernatant was centrifuged again for 45 min at 100,000 Â g. This supernatant containing approximately 200 mg protein in a volume of 60 ml was subject to an ammonium sulfate precipitation. An amount of ammonium sulfate gradually was added to obtain saturations of 30%. After resting for 14 h at 4 C, the sample was centrifuged at 12,000 Â g at 4 C for 30 min. The hydratase activity was detected in the supernatant fraction. This fraction was dialyzed against 20 mM Tris-HCl buffer pH 8.3. An amount of samples containing 5 mg of total protein was applied to a 4.6 Â 50 mm anion-exchange Q column (BioSuite Q-PEEK, Waters Corporation) mounted on an Ultimate 3000 HPLC system (Dionex). The column was pre-equilibrated with 20 mM Tris-HCl, pH 7.5 (Buffer A). The CA was eluted by a linear gradient from 0% Buffer A to 100% Buffer B (20 mM Tris-HCl, pH 7.5, containing 0.5 M NaCl) with a flow rate set to 1 ml/min with a continuous monitoring of the absorbance at 280 and 220 nm; 1-ml fractions were collected. The fractions were tested for the CA presence using the protonography technique. A peak of CA activity was eluted at a concentration of NaCl of approximately 0.1 M. At this stage of the purification, the CA was 60% pure and the obtained recovery was of 1 mg of the native protein. The active fraction was pooled, dialyzed against 20 mM Tris-HCl buffer pH 7.5 and further purified by affinity chromatography on a p-aminomethylbenzenesulfonamide agarose resin (pAMBS; Sigma-Aldrich). 1 ml of pAMBS resin was applied to an empty econocolumn (BioRad) and equilibrated with 0.1 M Tris-HCl, pH 7.5 buffer containing 0.2 M K 2 SO 4 , 0.5 mM EDTA. The sample containing about 1 mg of total protein was loaded on the p-AMBS column equilibrated as aforementioned. Unbound proteins were removed by washing extensively with the same buffer. The bound CA was eluted using 0.4 M KSCN dissolved in 0.1 M Tris-HCl, pH 7.5 buffer. The CA-containing fractions were pooled, dialyzed and concentrated by ultrafiltration. The CA-containing sample was subject to protonography. At this stage of the purification, the CA isolated from M. galloprovincialis was 90% pure with a total concentration of 0.1 mg. Its N-amino terminal sequence was determined for assigning the CA at one of the seven classes of CA described in literature.

SDS-PAGE, electroblotting and sequencing
Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) was carried out according to Laemmli 52 . Samples were dissolved in buffer with 5% b-mercaptoethanol. Gel was stained with Coomassie blue. Blotting from gel onto an Immobilion-P membrane was performed as described by Matsudaira 53 . N-terminal sequencing was performed on the blotted protein by automated Edman degradation 54 . Protein concentration was determined by Bio-Rad assay kit.

Carbonic anhydrase assay
CA activity assay was a medication of the procedure described by Capasso et al. 55 . Briefly, the assay was performed at 0 C using CO 2 as substrate following the pH variation due to the catalyzed conversion of CO 2 to bicarbonate. Bromothymol blue was used as the indicator of pH variation. The production of hydrogen ions during the CO 2 hydration reaction lowers the pH of the solution until the color transition point of the dye is reached. The time required for the color change is inversely related to the quantity of CA present in the sample. Wilbur-Anderson units were calculated according to the following definition: One Wilbur-Anderson unit (WAU) of activity is defined as (T0 À T)/T, where T0 (uncatalyzed reaction) and T (catalyzed reaction) are recorded as the time (in seconds) required for the pH to drop from 8.3 to the transition point of the dye in a control buffer and in the presence of enzyme, respectively.

Esterase activity
Activity for p-nitrophenylacetate (p-NpA) hydrolysis was determined at 0 C using a modification of the method proposed by Armstrong et al. 56 . The reaction mixture contained 0.3 ml of freshly prepared 3 mM p-NpA and 0.7 ml of 15 mM Tris sulfate buffer, pH 7.6. An aliquot of enzyme solution was added, and the catalyzed reaction was monitored reading the increase in A 348 nm for 5 min. The catalyzed reactions were corrected for the non-enzymatic reaction. One enzyme unit was defined as the amount capable of producing an OD 348 nm ¼ 0.03 in 5 min.

Kinetic and inhibition assays
An applied photophysics stopped-flow instrument has been used for assaying the CA catalyzed CO 2 hydration activity. Phenol red (at a concentration of 0.2 mM) has been used as indicator, working at the absorbance maximum of 557 nm, with 20 mM Tris (pH 8.3) as buffer, and 20 mM NaClO 4 (for maintaining constant the ionic strength), following the initial rates of the CA-catalyzed CO 2 hydration reaction for a period of 10-100 s. The CO 2 concentrations ranged from 1.7 to 17 mM for the determination of the kinetic parameters (by Lineweaver-Burk plots) and inhibition constants. For each inhibitor at least six traces of the initial 5-10% of the reaction have been used for determining the initial velocity. The uncatalyzed rates were determined in the same manner and subtracted from the total observed rates. Stock solutions of inhibitor (10-100 mM) were prepared in distilled-deionized water and dilutions up to 0.01 mM were done thereafter with the assay buffer. Inhibitor and enzyme solutions were preincubated together for 15 min at room temperature prior to assay, in order to allow for the formation of the E-I complex or for the eventual active site mediated hydrolysis of the inhibitor. The inhibition constants were obtained by non-linear least-squares methods using PRISM 3 and the Cheng-Prusoff equation, as reported earlier, and represent the mean from at least three different determinations.

Protonography
Protonography is a simple and inexpensive method, similar to zymography, which allowed the detection of CA activity on the polyacrylamide gel following the formation of H þ ions produced by hydratase activity of CAs [57][58][59] . Samples were run on SDS-PAGE at a concentration of about 1 lg per well. SDS-Page was carried out as described in the section "SDS-PAGE", with the exception that samples were mixed with Laemmli loading buffer without 2mercaptoethanol and without boiling the samples, in order to prevent protein denaturation induced by heating. The gel was run at 180 V until the dye front ran off the gel [57][58][59] . Following the electrophoresis, the gel was treated with Triton X-100 at 2.5% and kept under stirring for one hour to remove the SDS. The gel was subjected to a washing step of 20 min with 100 mM Tris-HCl, pH 8.3, containing 10% isopropanol. It was washed two times for 10 min. Finally, the gel was incubated for 30 min at 4 C under stirring with 0.1% bromothymol blue (BTB, the pH indicator) in 100 mM Tris-HCl pH 8.3. To detect the hydratase activity, the gel was immersed in distilled water saturated with CO 2 prepared by bubbling CO 2 into 200 ml of distilled water for about 3 h. The localized decrease of the pH value, due to the presence of the enzymatic activity of CAs, was detected through the formation of yellow band due to the change of color of the BTB from blue (alkaline pH) to yellow (acidic pH) 29,57-61 .

Results and discussion
Purification of the native form of CA The native CA was isolated and purified to homogeneity at 4 C from the mantles of about 30 mussels belonging to the species Mytilus galloprovincialis, the Mediterranean mussel. Most of the CA activity was recovered in the soluble fraction of cell extract after ultracentrifugation as described in "Materials and methods" section. Contaminants proteins, such as actin and chitinase, were precipitated by addition of solid ammonium sulfate (30% saturation). Following the ammonium sulfate precipitation, the hydratase activity was detected in the supernatant fraction, which was fractioned by anion-exchange chromatography as showed in Figure 1. The elution profile from anion-exchange chromatography showed the presence of at least 8 peaks with a significant absorption at 280 nm (Figure 1, panel A). All the peaks were analyzed by protonographic technique, which allowed the detection of CA activity on the polyacrylamide gel following the formation of H þ ions produced by hydratase activity of CAs. The protonogram was stained with bromothymol blue, an indicator monitoring the pH variation. This dye appears blue in its deprotonated form, while its color changes to yellow in the protonated form. The production of hydrogen ions during the CO 2 hydration lowers the pH of the solution until the transition point of the dye is reached (pH 6.8) and the yellow color appears on the gel (Figure 1, panel B). As shown in Figure 1, the protonograms clearly showed a hydratase activity only associated to the peak number 4 eluted at a concentration of NaCl of approximately 0.1 M and with a retention time of about 10 min. At this stage of the purification, the CA was 60% pure as in agreement with the SDS-PAGE analysis of the native enzyme (data not shown). The hydratase-containing peak was further purified by pAMBS affinity chromatography.

SDS-Page and protonography
In Figure 2 is shown the SDS-PAGEs obtained analyzing the sample after the tissue homogenization of the mussels (panel A) and the column affinity chromatography (panel B). The native CA was purified to an apparent homogeneity, as indicated by a single protein band after SDS-PAGE ( Figure 1, panel B, lane 3). The molecular weight estimated by SDS-PAGE was 50.0 kDA under both reducing and non-reducing conditions. Generally, a subunit molecular mass of about 26.0 kDa was calculated on the basis of the amino acid sequence translated from the aor b-CAs usually found in metazoans, although the Tridacna gigas CA was observed to possess a MW of 70 kDA, being a dual glycoprotein with two different CA domain, one at the amino-, the other at the carboxy-terminal parts, of the protein 51 . So, with the information available at this moment, we cannot rule out that the M. galloprovincialis CA may be a homodimer, or as the T. gigas enzyme, a glycoprotein with two different CA domains.
The hydratase activity of the purified mussel CA was confirmed by protonography. Figure 3 showed the protonogram obtained with the native mussel CA. The commercial bovine CA (bCA) was used as positive control. The yellow band corresponded to the CA position on the gel. The SDS-PAGE and protonogram showed clearly that the native mussel CA had a dimeric arrangement, suggesting that this enzyme may acts as a physiological dimer. This is very intriguing because generally a-CAs are monomeric enzymes. But, in the last years, it has been discovered dimeric arrangements for several members of the a-class, such as SspCA, from Sulfurihydrogenibium yellowstonense YO3AOP1 39 ; NgCA, from Neisseria gonorrhoeae 62 ; CrCA, from Chlamydomonas reinhardtii 63 ; AoCA, from Aspergillus oryzae 64 ; and three human isoforms, indicated with the acronyms hCA VI, hCA IX and hCA XII [65][66][67] . Moreover, CAs (a-type) from mammalian sources catalyze the reversible hydrolysis of esters. Thus, with p-NpA as substrate, the presence of esterase activity was investigated in the native mussel enzyme. The mussel CA showed an esterase specific activity 100 times smaller than that of the commercially available bovine bCA II.

Primary structure analysis
The N-amino terminal sequence of the electroblotted enzyme was found to be SWGYGNDNGP. Using the Basic Local Alignment Search Tool (BLAST) and using as query sequence the N-amino terminal obtained as described above, the purified native CA completely matched (query cover 100%) the N-amino terminal of the a-CA previously identified in the genome of the Mediterranean mussel, Mytilus galloprovincialis (Figure 4). The M. galloprovincialis CA is a polypeptide chain formed by 255 amino acid residues. In Figure 5, the M. galloprovincialis CA sequence was aligned with Homo sapiens CA isoforms (hCA I and hCA II) sequences. It may be   observed that similar to the other investigated a-CA, the mussel metalloenzyme has the conserved three His ligands, which coordinate the Zn(II) ion crucial for catalysis (His94, 96 and 119, hCA I numbering system). Interesting to note that the proton shuttle residue (His64) is conserved in the human isoforms, but is missing in the mussel enzyme, being replaced by a Lys residue. This is also the situation for the human isoform hCA III, which has Lys64 instead of His64, and also shows a reduced catalytic activity compared with isoforms hCA I and II 68 . This residue assists the rate determining step of the catalytic cycle transferring a proton from the water coordinated to the Zn(II) ion to the environment, with formation of zinc hydroxide representing the nucleophilic species of the enzyme. A Lys in position 64 is less efficient than a His for assisting this process, which explains why the molluskan enzyme has a lower catalytic activity compared with hCA I and II 68 . M. galloprovincialis enzyme also has the gate-keeping residues (Glu106 and Thr199), which orientate the substrate for catalysis, and are also involved in the binding of inhibitors.

Enzyme kinetics
Using the stopped-flow technique, the kinetic parameters were determined for the native a-CA using CO 2 as a substrate 69 . The activity of the mussel enzyme was compared with that of other a-CAs such as Homo sapiens isoforms (hCA I and hCA II), Stylophora pistillata isoforms (STPCA and STPCA-2) ( Table 1) 18,70,71 . As shown in Table 1, the mussel a-CA had a k cat ¼ 4.1 Â 10 5 s À1 with a k cat / K m ratio 1.38 times lower with respect to that of the hCA I, which possessed a k cat of 2.0 Â 10 5 s À1 . Moreover, the kinetic constants of the native mussel CA are very similar to those obtained for the coral isoform STPC and only slightly lower than the isoform STPCA-2. This is very intriguing since mussel a-CA respect to the two coral isoforms, lacked of a His64. This residue is involved in the transfer of proton from the water coordinated to the Zn(II) ion to the environment with the function to accelerate the rate of the catalytic cycle. STPCA-2 from scleractinian coral Stylophora pistillata is considered highly active and has an intracellular localization in the oral endoderm and in the aboral tissue of the coral. Interesting to note that the native mussel CA is not well affected by acetazolamide inhibition showing a K I of 380 mM ( Table 1). The intracellular localization of the native mussel CA and its kinetic similarity with STPCA-2, could allow that the Mytilus galloprovincialis a-CA is involved in pH regulation and/or in the delivery of bicarbonate helping the mussel during the formation of the shell.

Conclusions
A a-CA has been purified and characterized biochemically from the mollusk Mytilus galloprovincialis. As in most mollusks, this a-CA  Errors in the range of ±5% of the reported data from three different assays.
seems to be involved in the biomineralization processes leading to the precipitation of calcium carbonate in the mussel shell. The new enzyme has a molecular weight of 50 kDa, which is roughly two times higher than that of a monomeric a-class enzyme. Thus, Mytilus galloprovincialis a-CA is either a dimer, or similar to the Tridacna gigas CA described earlier, may have two different CA domains in its polypeptide chain. The Mytilus galloprovincialis a-CA sequence contains the three His residues acting as zinc ligands and the gate-keeper residues (Glu106-Thr199) present in all a-CAs, but had a Lys in position 64 and not a His as proton shuttling residue, being thus similar to the human isoform hCA III. This probably explains the relatively low catalytic activity of Mytilus galloprovincialis a-CA, with the following kinetic parameters for the CO 2 hydration reaction: k cat ¼ 4.1 Â 10 5 s À1 and k cat /K m of 3.6 Â 10 7 M À1 Â s À1 . The enzyme activity was poorly inhibited by the sulfonamide acetazolamide, with a K I of 380 nM. This study is one of the few describing in detail the biochemical characterization of a molluskan CA and may be useful for understanding in better detail the phylogeny of these enzymes and their role in biocalcification processes. In addition, the mussel CA is an attractive candidate for its potential use in carbon dioxide sequestration. The recombinant DNA technology will make possible the ingegnerization of the mussel CA and production of large quantities of the metalloenzyme with the aim to use the M. galloprovincialis CA, either free or immobilized, in the CO 2 biomimetic capture process.
Disclosure statement